*** Publication

These are the replication files for the following publication:

Jelena Dzankic, Mira Seyfettinoglu, Ayelet Shachar, Maarten Vink, and Luuk van der Baaren, The Archipelago Capitalism of Citizenship-by-Investment. Comparative Political Studies.

Replication files at Harvard Dataverse at: https://doi.org/10.7910/DVN/KJXQVM.

*** Abstract

Citizenship-by-investment (CBI) programs, granting citizenship in return for financial payment or investment, have become a global phenomenon in recent years. The workings of these exceptional programs have caused controversy in real-life politics, ranging from protests, to the downfall of politicians, and to punitive bilateral and international measures. Even so, knowledge on why countries would put their citizenship up for sale has remained limited. This study combines insights from political science and legal theory to develop an original approach to understand states' propensity to adopt investor citizenship policies as part of the offshore world, or the legal spaces of ‘archipelago capitalism’. We leverage a novel global longitudinal CBI dataset (1960 to 2023) to probe the empirical plausibility of this argument. In line with our expectations, we find that microstates, middle-income countries, and tax havens are more likely to implement CBI programs. CBI supply reflects a contemporary form of small state ingenuity. 

*** Technical requirements

The replication files require the statistical software R to run (version 4.5.1 or higher, available here: https://www.r-project.org). Analyses performed in RStudio Version 2025.05.1+513.

*** Instructions

To run the replication, follow these four steps:

1. Create a new project directory in RStudio 

2. Download the following files from the Dataverse and save these into the project directory:
- data_preparation.R
- descriptive_trends.R
- models.R
- cbi_data.xlsx
- dat_islands_legal.Rdata
- dat_sids.Rdata
- imf_dataset_gdpcapita_2025-07-23.csv

3. Run the code 'data_preparation.R'

Running this code reconstructs four data files, which will be saved into your R project directory in a new sub-folder '01_data': 
- cbi_data_full.csv [this is the data file used for the descriptive trend plots Figures 1 and 2]
- cbi_dat_imp_cens.csv [this is the data file used for the main models reported in SM2, SM3, SM6, SM7, SM9, SM10(1), SM11, SM12, SM13, SM14]
- cbi_dat_cens.csv [this is the data file used for the robustness check reported in SM10(2)]
- cbi_dat_imp_generic_cens.csv [this is the data file used for the robustness check reported in SM8]

4. Run the code 'descriptive_trends.R'.

Running this code produces the descriptive trend plots Figure 1 and Figure 2, and the descriptive tables reported in SM3 and SM4.

Tables and figures will be stored in the project directory in a new sub-folder '02_plots_tables'


5. Run the code 'models.R'.

Running this code produces the output of the analysis as reported in the paper and the supplementary materials. 

Tables and figures will be stored in the project directory in sub-folder '02_plots_tables'


*** Contact

For questions about the replication materials, contact Maarten Vink at maarten.vink@eui.eu

Date: 14/10/2025


***END***

